Study on Computer Science Undergraduate Students Dropout at the University of Brasilia.

Dropout is a chronic problem that affects education in Brazil at all levels. When considering higher education, Brazil has one of the highest university dropout rates in public and private institutions. Such a problem is considered a social loss as well as a misuse of resources. This paper aims to identify academic and social performance factors that influence the dropout of undergraduate students from Computer Science major at the University of Brasilia (UnB). The data set consists of 879 observations and 16 variables with aspects of the students enrolled in Computer Science from the first semester of 2014 to the second semester of 2019. The survival analysis methodology was employed, more specifically, the Log-Normal regression model, in which the response variable was the time (number of semesters) until dropout (known as failure time) or the last follow-up of the student (either the completion of the course or the student is still enrolled by 2019/2). The model proved to be robust in the residual analysis and in presenting results consistent with the dropout literature. This work can assist in the development of policies and educational strategies to decrease the number of dropouts in the computer science undergraduate major.